Spatio-Temporal Trajectory Gatherings Pattern Mining Method Based on R* Tree Index

Spatio-temporal traj ectory gatherings pattern refers to a group of moving objects getting together and forming durable and stable areas in high density, which is often used to monitor and predict abnormal group events. Yet it is still a challenge to discover the specific gathering pattern effectively and efficiently with the generation of large volume of trajectory data. In the light, we propose a novel improved algorithm based on R* tree index for mining gatherings pattern of spatio-temporal trajectory, in which we consider moving objects' direction property in the trajectory simplify stage and we use R* tree index in the clustering stage to improve the effectiveness and efficiency. The experiment based on the real trajectory dataset demonstrates that the proposed method is superior in efficiency to the existing method and can guarantee the same effectiveness for moving object gatherings pattern mining.